sbcli-pre 25.5.3__tar.gz → 25.5.5__tar.gz

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-pre
-Version: 25.5.3
+Version: 25.5.5
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/sbcli_pre.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-pre
-Version: 25.5.3
+Version: 25.5.5
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/sbcli_pre.egg-info/entry_points.txt

@@ -1,3 +1,4 @@
 [console_scripts]
 sbcli-pre = simplyblock_cli.cli:main
+sbctl = simplyblock_cli.cli:main
 

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/setup.py

@@ -97,7 +97,7 @@ setup(
     entry_points={
         'console_scripts': [
             f'{COMMAND_NAME}=simplyblock_cli.cli:main',
-            # 'sbctl=simplyblock_cli.cli:main',
+            'sbctl=simplyblock_cli.cli:main',
         ]
     },
     include_package_data=True,

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_cli/cli.py

@@ -100,7 +100,7 @@ class CLIWrapper(CLIWrapperBase):
     def init_storage_node__add_node(self, subparser):
         subcommand = self.add_sub_command(subparser, 'add-node', 'Adds a storage node by its IP address')
         subcommand.add_argument('cluster_id', help='Cluster id', type=str)
-        subcommand.add_argument('node_ip', help='IP of storage node to add', type=str)
+        subcommand.add_argument('node_addr', help='Address of storage node api to add, like <node-ip>:5000', type=str)
         subcommand.add_argument('ifname', help='Management interface name', type=str)
         argument = subcommand.add_argument('--journal-partition', help='1: auto-create small partitions for journal on nvme devices. 0: use a separate (the smallest) nvme device of the node for journal. The journal needs a maximum of 3 percent of total available raw disk space.', type=int, default=1, dest='partitions')
         if self.developer_mode:

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_cli/clibase.py

@@ -115,7 +115,7 @@ class CLIWrapperBase:
 
     def storage_node__add_node(self, sub_command, args):
         cluster_id = args.cluster_id
-        node_ip = args.node_ip
+        node_addr = args.node_addr
         ifname = args.ifname
         data_nics = args.data_nics
         spdk_image = args.spdk_image
@@ -134,7 +134,7 @@ class CLIWrapperBase:
 
         out = storage_ops.add_node(
             cluster_id=cluster_id,
-            node_ip=node_ip,
+            node_addr=node_addr,
             iface_name=ifname,
             data_nics_list=data_nics,
             max_snap=max_snap,

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/cluster_ops.py

@@ -683,6 +683,31 @@ def cluster_activate(cl_id, force=False, force_lvstore_create=False):
                 set_cluster_status(cl_id, ols_status)
                 return False
 
+    snodes = db_controller.get_storage_nodes_by_cluster_id(cl_id)
+    for snode in snodes:
+        if snode.status != StorageNode.STATUS_ONLINE:
+            continue
+
+        if not snode.is_secondary_node:
+            continue
+
+        logger.info(f"recreating secondary node {snode.get_id()}")
+        ret = storage_node_ops.recreate_lvstore_on_sec(snode)
+
+        snode = db_controller.get_storage_node_by_id(snode.get_id())
+        if ret:
+            snode.lvstore_status = "ready"
+            snode.write_to_db()
+
+        else:
+            snode.lvstore_status = "failed"
+            snode.write_to_db()
+            logger.error(f"Failed to restore lvstore on node {snode.get_id()}")
+            if not force:
+                logger.error("Failed to activate cluster")
+                set_cluster_status(cl_id, ols_status)
+                return False
+
 
     if not cluster.cluster_max_size:
         cluster = db_controller.get_cluster_by_id(cl_id)

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/controllers/tasks_controller.py

@@ -179,13 +179,17 @@ def get_active_dev_restart_task(cluster_id, device_id):
     return False
 
 
-def get_active_node_mig_task(cluster_id, node_id):
+def get_active_node_mig_task(cluster_id, node_id, distr_name=None):
     tasks = db.get_job_tasks(cluster_id)
     for task in tasks:
         if task.function_name in [JobSchedule.FN_FAILED_DEV_MIG, JobSchedule.FN_DEV_MIG,
                                   JobSchedule.FN_NEW_DEV_MIG] and task.node_id == node_id:
             if task.status == JobSchedule.STATUS_RUNNING and task.canceled is False:
-                return task.uuid
+                if distr_name:
+                    if "distr_name" in task.function_params and task.function_params["distr_name"] == distr_name:
+                        return task.uuid
+                else:
+                    return task.uuid
     return False
 
 

sbcli-pre-25.5.5/simplyblock_core/env_var

@@ -0,0 +1,6 @@
+SIMPLY_BLOCK_COMMAND_NAME=sbcli-pre
+SIMPLY_BLOCK_VERSION=25.5.5
+
+
+SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:R25.5-PRE
+SIMPLY_BLOCK_SPDK_ULTRA_IMAGE=simplyblock/spdk:main-latest

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/mgmt_node_ops.py

@@ -199,7 +199,7 @@ def remove_mgmt_node(uuid):
 
     logger.info("Leaving swarm...")
     node_docker = docker.DockerClient(base_url=f"tcp://{snode.docker_ip_port}", version="auto")
-    node_docker.swarm.leave()
+    node_docker.swarm.leave(force=True)
 
     mgmt_events.mgmt_remove(snode)
     logging.info("done")

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/models/storage_node.py

@@ -146,7 +146,7 @@ class StorageNode(BaseNodeObject):
             #
             # raise
 
-    def create_hublvol(self, cluster_nqn):
+    def create_hublvol(self):
         """Create a hublvol for this node's lvstore
         """
         logger.info(f'Creating hublvol on {self.get_id()}')
@@ -159,7 +159,7 @@ class StorageNode(BaseNodeObject):
                 raise RPCException('Failed to create hublvol')
             self.hublvol = HubLVol({
                 'uuid': hublvol_uuid,
-                'nqn': f'{cluster_nqn}:lvol:{hublvol_uuid}',
+                'nqn': f'{self.host_nqn}:lvol:{hublvol_uuid}',
                 'bdev_name': f'{self.lvstore}/hublvol',
                 'model_number': str(uuid4()),
                 'nguid': utils.generate_hex_string(16),
@@ -212,6 +212,13 @@ class StorageNode(BaseNodeObject):
                 return True
             except RPCException:
                 pass
+        else:
+            try:
+                self.create_hublvol()
+                return True
+            except RPCException as e:
+                logger.error("Error establishing hublvol: %s", e.message)
+                # return False
 
         return self.hublvol
 

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/rpc_client.py

@@ -148,8 +148,8 @@ class RPCClient:
             "max_io_size": 131072,
             "io_unit_size": 8192,
             "max_aq_depth": 128,
-            "num_shared_buffers": 30000,
-            "buf_cache_size": 1024,
+            "num_shared_buffers": 24576,
+            "buf_cache_size": 512,
             "dif_insert_or_strip": False,
             "c2h_success": True,
             "sock_priority": 0

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/services/health_check_service.py

@@ -197,7 +197,7 @@ while True:
                             connected_jms.append(remote_device.get_id())
 
                     for jm_id in snode.jm_ids:
-                        if jm_id not in connected_jms:
+                        if jm_id and jm_id not in connected_jms:
                             for nd in db.get_storage_nodes():
                                 if nd.jm_device and nd.jm_device.get_id() == jm_id:
                                     if nd.status == StorageNode.STATUS_ONLINE:

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/services/tasks_runner_failed_migration.py

@@ -138,7 +138,8 @@ while True:
             for task in tasks:
                 if task.function_name == JobSchedule.FN_FAILED_DEV_MIG:
                     if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
-                        active_task = tasks_controller.get_active_node_mig_task(task.cluster_id, task.node_id)
+                        active_task = tasks_controller.get_active_node_mig_task(
+                            task.cluster_id, task.node_id, task.function_params["distr_name"])
                         if active_task:
                             logger.info("task found on same node, retry")
                             continue

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/services/tasks_runner_migration.py

@@ -134,7 +134,8 @@ while True:
                 if task.function_name == JobSchedule.FN_DEV_MIG and task.status != JobSchedule.STATUS_DONE:
                     task = db.get_task_by_id(task.uuid)
                     if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
-                        active_task = tasks_controller.get_active_node_mig_task(task.cluster_id, task.node_id)
+                        active_task = tasks_controller.get_active_node_mig_task(
+                            task.cluster_id, task.node_id,  task.function_params["distr_name"])
                         if active_task:
                             logger.info("task found on same node, retry")
                             continue

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/services/tasks_runner_new_dev_migration.py

@@ -149,7 +149,8 @@ while True:
             for task in tasks:
                 if task.function_name == JobSchedule.FN_NEW_DEV_MIG:
                     if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
-                        active_task = tasks_controller.get_active_node_mig_task(task.cluster_id, task.node_id)
+                        active_task = tasks_controller.get_active_node_mig_task(
+                            task.cluster_id, task.node_id,  task.function_params["distr_name"])
                         if active_task:
                             logger.info("task found on same node, retry")
                             continue

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_core/storage_node_ops.py

@@ -778,13 +778,13 @@ def _connect_to_remote_jm_devs(this_node, jm_ids=None):
     return new_devs
 
 
-def add_node(cluster_id, node_ip, iface_name, data_nics_list,
+def add_node(cluster_id, node_addr, iface_name, data_nics_list,
              max_snap, spdk_image=None, spdk_debug=False,
              small_bufsize=0, large_bufsize=0,
              num_partitions_per_dev=0, jm_percent=0, enable_test_device=False,
              namespace=None, enable_ha_jm=False, id_device_by_nqn=False,
              partition_size="", ha_jm_count=3, full_page_unmap=False):
-    snode_api = SNodeClient(node_ip)
+    snode_api = SNodeClient(node_addr)
     node_info, _ = snode_api.info()
     if node_info.get("nodes_config") and node_info["nodes_config"].get("nodes"):
         nodes = node_info["nodes_config"]["nodes"]
@@ -804,7 +804,7 @@ def add_node(cluster_id, node_ip, iface_name, data_nics_list,
             logger.error("Cluster not found: %s", cluster_id)
             return False
 
-        logger.info(f"Adding Storage node: {node_ip}")
+        logger.info(f"Adding Storage node: {node_addr}")
 
         if not node_info:
             logger.error("SNode API is not reachable")
@@ -908,7 +908,7 @@ def add_node(cluster_id, node_ip, iface_name, data_nics_list,
         if ssd_pcie:
             for ssd in ssd_pcie:
                 for node in db_controller.get_storage_nodes_by_cluster_id(cluster_id):
-                    if node.api_endpoint == node_ip:
+                    if node.api_endpoint == node_addr:
                         if ssd in node.ssd_pcie:
                             logger.error(f"SSD is being used by other node, ssd: {ssd}, node: {node.get_id()}")
                             return False
@@ -935,7 +935,7 @@ def add_node(cluster_id, node_ip, iface_name, data_nics_list,
 
         total_mem = minimum_hp_memory
         for n in db_controller.get_storage_nodes_by_cluster_id(cluster_id):
-            if n.api_endpoint == node_ip:
+            if n.api_endpoint == node_addr:
                 total_mem += n.spdk_mem
         total_mem += utils.parse_size("500m")
         logger.info("Deploying SPDK")
@@ -999,7 +999,7 @@ def add_node(cluster_id, node_ip, iface_name, data_nics_list,
         snode.rpc_username = rpc_user
         snode.rpc_password = rpc_pass
         snode.cluster_id = cluster_id
-        snode.api_endpoint = node_ip
+        snode.api_endpoint = node_addr
         snode.host_secret = utils.generate_string(20)
         snode.ctrl_secret = utils.generate_string(20)
         snode.number_of_distribs = number_of_distribs
@@ -1761,6 +1761,15 @@ def restart_storage_node(
             if devs:
                 dev = db_controller.get_jm_device_by_id(devs[0])
                 snode.remote_jm_devices.append(dev)
+        if not snode.jm_ids:
+            snode.jm_ids = []
+            for rem_jm in snode.remote_jm_devices:
+                if rem_jm.get_id():
+                    snode.jm_ids.append(rem_jm.get_id())
+
+            snode.jm_ids = snode.jm_ids[:constants.HA_JM_COUNT-1]
+            snode.write_to_db(db_controller.kv_store)
+
         snode.remote_jm_devices = _connect_to_remote_jm_devs(snode)
     snode.health_check = True
     snode.lvstore_status = ""
@@ -3088,30 +3097,37 @@ def add_lvol_thread(lvol, snode, lvol_ana_state="optimized"):
 def get_sorted_ha_jms(current_node):
     db_controller = DBController()
     jm_count = {}
-    mgmt_ips = []
+    jm_dev_to_mgmt_ip = {}
+
     for node in db_controller.get_storage_nodes_by_cluster_id(current_node.cluster_id):
-        if (node.get_id() == current_node.get_id() or node.status != StorageNode.STATUS_ONLINE):  # pass
-            continue
-        if node.mgmt_ip == current_node.mgmt_ip:
-            continue
-        if node.mgmt_ip in mgmt_ips:
+        if node.get_id() == current_node.get_id():  # pass
             continue
 
-        if node.jm_device and node.jm_device.status == JMDevice.STATUS_ONLINE:
-            jm_count[node.jm_device.get_id()] = 1
-            mgmt_ips.append(node.mgmt_ip)
+        if node.jm_device and node.jm_device.status == JMDevice.STATUS_ONLINE and node.jm_device.get_id():
+            jm_count[node.jm_device.get_id()] = 0
+            jm_dev_to_mgmt_ip[node.jm_device.get_id()] = node.mgmt_ip
 
     for node in db_controller.get_storage_nodes_by_cluster_id(current_node.cluster_id):
-        if (node.get_id() == current_node.get_id() or node.status != StorageNode.STATUS_ONLINE):  # pass
+        if node.get_id() == current_node.get_id():  # pass
             continue
-        if node.mgmt_ip == current_node.mgmt_ip or not node.jm_ids:
+        if not node.jm_ids:
             continue
         for rem_jm_id in node.jm_ids:
             if rem_jm_id in jm_count:
                 jm_count[rem_jm_id] += 1
 
+    mgmt_ips = []
     jm_count = dict(sorted(jm_count.items(), key=lambda x: x[1]))
-    return list(jm_count.keys())[:3]
+    out = []
+    for jm_id in jm_count.keys():
+        if jm_id:
+            if jm_dev_to_mgmt_ip[jm_id] in mgmt_ips:
+                continue
+            if jm_dev_to_mgmt_ip[jm_id] == current_node.mgmt_ip:
+                continue
+            mgmt_ips.append(jm_dev_to_mgmt_ip[jm_id])
+            out.append(jm_count)
+    return out[:constants.HA_JM_COUNT-1]
 
 
 def get_node_jm_names(current_node, remote_node=None):
@@ -3126,6 +3142,9 @@ def get_node_jm_names(current_node, remote_node=None):
 
     if current_node.enable_ha_jm:
         for jm_id in current_node.jm_ids:
+            if not jm_id:
+                continue
+
             if remote_node:
                 if remote_node.jm_device.get_id() == jm_id:
                     jm_list.append(remote_node.jm_device.jm_bdev)
@@ -3298,9 +3317,8 @@ def create_lvstore(snode, ndcs, npcs, distr_bs, distr_chunk_bs, page_size_in_blo
         sec_rpc_client.bdev_examine(snode.raid)
         sec_rpc_client.bdev_wait_for_examine()
 
-        cluster_nqn = db_controller.get_cluster_by_id(snode.cluster_id).nqn
         try:
-            snode.create_hublvol(cluster_nqn)
+            snode.create_hublvol()
 
         except RPCException as e:
             logger.error("Error establishing hublvol: %s", e.message)

{sbcli-pre-25.5.3 → sbcli-pre-25.5.5}/simplyblock_web/blueprints/web_api_storage_node.py

@@ -182,14 +182,14 @@ def storage_node_add():
     if 'cluster_id' not in req_data:
         return utils.get_response_error("missing required param: cluster_id", 400)
 
-    if 'node_ip' not in req_data:
-        return utils.get_response_error("missing required param: node_ip", 400)
+    if 'node_addr' not in req_data:
+        return utils.get_response_error("missing required param: node_addr", 400)
 
     if 'ifname' not in req_data:
         return utils.get_response_error("missing required param: ifname", 400)
 
     cluster_id = req_data['cluster_id']
-    node_ip = req_data['node_ip']
+    node_addr = req_data['node_addr']
     ifname = req_data['ifname']
     max_snap = int(req_data.get('max_snap', 500))
 
@@ -201,9 +201,9 @@ def storage_node_add():
     enable_test_device = False
     param = req_data.get('enable_test_device')
     if param:
-        if type(param) == bool:
+        if isinstance(param, bool):
             enable_test_device = param
-        elif type(param) == str:
+        elif isinstance(param, str):
             enable_test_device = param == "true"
 
     spdk_image = None
@@ -245,7 +245,7 @@ def storage_node_add():
 
     tasks_controller.add_node_add_task(cluster_id, {
         "cluster_id": cluster_id,
-        "node_ip": node_ip,
+        "node_addr": node_addr,
         "iface_name": ifname,
         "data_nics_list": data_nics,
         "max_snap": max_snap,

sbcli-pre-25.5.3/simplyblock_core/env_var

@@ -1,6 +0,0 @@
-SIMPLY_BLOCK_COMMAND_NAME=sbcli-pre
-SIMPLY_BLOCK_VERSION=25.5.3
-
-
-SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:R25.3-PRE-Hotfix-upgrade
-SIMPLY_BLOCK_SPDK_ULTRA_IMAGE=simplyblock/spdk:main-latest